KR102070911B1 - Method for identifying errors occurred by massively parallel sequencing and an apparatus for the same - Google Patents

Abstract

(a) selecting at least one first read to be verified among reads generated by superparallel sequencing; (b) recovering the nucleic acid fragment corresponding to the first read from the sequencing plate; (c) performing sequencing on each of the fragments or amplification products thereof to generate a second read; And (d) comparing the second read with the corresponding first read to verify the error of the first read. In addition, there is provided an error checking apparatus for super parallel sequencing for performing the above method.

Description

Method for identifying errors occurred by massively parallel sequencing and an apparatus for the same}

The present invention relates to an error checking method and apparatus for hyperparallel sequencing.

The biggest disadvantage of next-generation sequencing is the high error rate. Compared with the Sanger sequencing method, which is a conventional sequencing method, it has a limit of about 10 times higher probability of erroneous reading of the sequencing at a probability of 0.1% to 1% per base. In fact, if you want to analyze the minor allele frequency (MAF) of less than 0.1%, it is impossible to distinguish between the analysis error on the next-generation sequencing system and the actual MAF molecular variation. The scope of the sequence analysis is limited.

Sequencing errors that occur in next-generation sequencing are mostly system errors that occur after detecting light from a sample and matching the corresponding sequencing. There is little error. Therefore, a more precise analysis method is required to distinguish between errors in the sequencing system or molecular variation of the actual gene.

One aspect provides an error checking method of hyperparallel sequencing.

Another aspect provides an error checking device for hyperparallel sequencing.

One aspect includes (a) selecting at least one first read to be verified among reads generated by superparallel sequencing; (b) recovering the nucleic acid fragment corresponding to the first read from the sequencing plate; (c) performing sequencing on each of the fragments or amplification products thereof to generate a second read; And (d) comparing the second read with a corresponding first read to verify an error of the first read.

1 is a process flow diagram illustrating an embodiment of a method for error checking of hyperparallel sequencing. Referring to FIG. 1, in step 110, at least one first read to be verified among the reads generated by the super parallel sequencing is selected.

The super parallel sequencing may also be referred to as next generation sequencing (NGS) or high-throughput sequencing. The hyperparallel sequencing includes a method currently known as next generation sequencing and a method that can be developed in the future. The hyperparallel sequencing consists of sequencing by synthesis, ion-torrent sequencing, pyrosequencing, ligation sequencing, nanopore sequencing, and single-molecule real-time sequencing. It is selected from, but is not limited thereto.

As used herein, the term "read" refers to one nucleic acid fragment analyzed by the next generation sequencing method. At least one of the first reads may be a read comprising or presumed to contain a mutated sequence based on the reference sequence. The reference sequence may be sequence information stored in a known sequence database including human genome sequence information. The reference sequence may be a sequence of nucleic acid molecules that are aligned together in a hyperparallel sequencing analysis. In addition, all or part of the first read may include the same sequence as the reference sequence. That is, all or some of the one or more first reads may be reads that do not include any of the sequences and mutated sequences generated due to sequencing errors relative to the reference sequence.

In step 120, the nucleic acid fragment corresponding to the first read is recovered from the sequencing plate.

The step 120 may be performed by hybridization, amplification by PCR, chemical separation, physical separation, or a combination thereof. In the case of amplification by hybridization or PCR, a desired nucleic acid fragment can be recovered from the sequencing plate by capturing or amplifying by using a probe or primer comprising at least some of the sequences of the fragment to be recovered or a sequence complementary thereto. In the case of the chemical separation, the desired fragment can be recovered by breaking the bond between the fragment to be recovered and the sequencing plate or by separating a portion of the fragment to be recovered.

The physical separation may be to recover the nucleic acid fragments by applying energy or leaving a portion of the support or sequencing plate to which the fragments are bound by magnetic material. The physical separation may be performed by using the position information of the fragment to be recovered in the sequencing plate, without amplification by hybridization or PCR. The physical separation method is, for example, a method called sniper cloning or a pulse lazer optical retrieval system, a method disclosed in JP-A-2014-0075611, or a method modified therefrom. It may be to use. The support may be a solid support, for example beads, to which nucleic acid fragments are bound. The energy application may be made through laser irradiation.

In step 120, the fragments may be recovered in a space separated from each other. Each fragment may be separated and recovered in each well of a storage vessel having a plurality of partitioned wells, such as a sequencing plate.

In step 120 the fragments can be recovered as a mixture of fragments to which a unique identifier has been added for each fragment. The unique identification material serves to index a specific fragment of a mixture consisting of a plurality of fragments, and may also be referred to as a barcode or a tag. The substance may be a biochemical molecule comprising a nucleic acid, a polypeptide, a peptide, or a combination thereof. To this end, a step of adding a unique identifier for each nucleic acid fragment to be recovered before step 120 may be further included. The addition of the unique identifier may be performed by amplification, ligation or peptide binding by PCR.

In step 130, sequencing of each fragment or amplification product thereof is performed to generate a second read.

The step 130 may be performed in a state where each of the fragments or their amplification products are mixed or separated. For example, when the fragment is recovered as a mixture of fragments to which a unique identifier is added for each fragment in step 120, the amplification reaction and / or sequencing may be performed in the mixed state. In addition, when each fragment is recovered in a space separated from each other in step 120, the amplification reaction and / or sequencing for each fragment may be made in a space separated.

The sequencing can be accomplished by methods including hyperparallel sequencing, Sanger sequencing, mass spectrometry, electrophoresis, hybridization, digital PCR, allele-specific PCR, quantitative PCR, fluorescence based sorting, or a combination thereof.

In operation 140, an error of the first lead is verified by comparing the second lead with a corresponding first lead.

The corresponding first lead is connected to the second lead when the fragments are separated from each other in space, for example, in each well of a storage vessel having a plurality of partitioned wells. Corresponding fragments may be matched by identifying the location information of the stored wells. Alternatively, the corresponding first read may be matched using a unique identification sequence added to the second read. The comparison of the reads can be performed using sequence alignment tools known to those skilled in the art or tools developed for read sequence alignment. The sequence alignment tool may be, for example, but not limited to, BWA, BarraCUDA, BBMap, BLASTN, Bowtie, NextGENe, or UGENE.

If the ratio of the reads having the same sequence as the first read in the corresponding region with the first read among the second reads corresponding to the first read is less than a predetermined value, the corresponding region of the first read may be caused by a sequencing error. It can be judged that. The constant value may be determined according to the sequence to be analyzed or other purposes. The predetermined value may be, for example, 50% to 95%, 55% to 95%, 60% to 90%, 65% to 90%, 70% to 85%, or 75% to 80%. In addition, when the ratio is greater than or equal to a certain value, it may be determined that the corresponding site of the first read is a sequence actually present in the nucleic acid fragment corresponding to the first read.

Another aspect includes a read selector configured to select at least one first read to be verified among reads generated by hyperparallel sequencing; A nucleic acid recovery unit for recovering a nucleic acid fragment corresponding to the first read from a sequencing plate; A lead generation unit generating a second lead through sequencing of each fragment or amplification product thereof; And a discriminating unit comparing the second read with the first read corresponding thereto and verifying an error of the first read.

2 is a diagram illustrating an embodiment of an error checking apparatus for hyperparallel sequencing. Referring to FIG. 2, the error checking apparatus 600 for super parallel sequencing includes a read selector 610 for selecting at least one first read to be verified among reads generated by super parallel sequencing; A nucleic acid recovery unit 620 for recovering a nucleic acid fragment corresponding to the first read from a sequencing plate; A read generator 630 for generating a second read through sequencing of each fragment or an amplification product thereof; And a determination unit 640 for comparing an error of the first lead by comparing the second lead with a first lead corresponding thereto.

In the read selector 610, at least one of the first reads may be a read including a mutated sequence based on a reference sequence. At least one of the first reads may be a read that is assumed to contain a mutated sequence relative to the reference sequence. In addition, all or part of the first read may include the same sequence as the reference sequence.

The nucleic acid recovery unit 620 may include a nucleic acid reaction unit that adds a unique identification material to each fragment through a hybridization reaction or an amplification reaction by PCR. The unique identification material is as described above. In one embodiment, the nucleic acid reaction unit may be provided with a storage means and a supply means for the reagent used in the hybridization reaction or amplification reaction by PCR. In one embodiment, the nucleic acid reaction unit may include a temperature control device for controlling the reaction temperature. Alternatively, in one embodiment, the nucleic acid reaction unit may include a reaction promoting device for applying a physical force such as agitation or vibration in the space in which the reaction occurs.

In addition, the nucleic acid recovery unit 620 may include an energy applying unit for applying energy to detach a portion of the support or the sequencing plate to which the nucleic acid fragment corresponding to the first read is bound from the plate. The nucleic acid recovery unit 620 may include a laser light source, an imaging device, and a light collecting device. The imaging device is for reading the position of the nucleic acid fragment to be recovered, and may be an optical microscope or an electron microscope. In one embodiment, the nucleic acid recovery unit 620 may include the above-described pulse laser optical recovery system. The nucleic acid recovery unit 620 may include a storage container for recovering each of the fragments in a space separated from each other.

The read generator 630 may include an apparatus for performing ultra-parallel sequencing, Sanger sequencing, mass spectrometry, electrophoresis, hybridization, digital PCR, allele-specific PCR, quantitative PCR, fluorescence-based classification, or a combination thereof. Can be.

The determination unit 640 may sequence the corresponding portions of the first reads according to the ratio of the reads having the same sequence as the first reads in the corresponding portions of the second reads corresponding to the first reads. You can determine whether there is an error. The determination unit 640 may include a sequence alignment tool or a tool developed for read sequence alignment. A detailed error verification method of the determination unit 640 is as described above in the error checking method of the super parallel sequencing.

According to an error checking method or apparatus for hyperparallel sequencing according to an aspect, an error sequence generated in superparallel sequencing can be accurately and easily identified. According to the method or apparatus, it is possible to distinguish with high accuracy the error sequence generated in superparallel sequencing and the mutant sequence which appears at a very low frequency. In addition, according to the method or apparatus, it is possible to accurately and quickly identify errors in superparallel sequencing or to identify mutant sequences that appear at a very low frequency without adding an identifying substance to the nucleic acid.

1 is a process flow diagram illustrating an embodiment of a method for error checking of hyperparallel sequencing.
2 is a diagram illustrating an embodiment of an error checking apparatus for hyperparallel sequencing.
Figure 3 is a process flow diagram showing an embodiment of the error analysis method of hyper-parallel sequencing.
4 shows some results of barcode PCR.
5 is a visualized representation of the BWA alignment results.
6A and 6B show graphs comparing sequencing results by barcodes.

Hereinafter, the present invention will be described in more detail with reference to the following examples. However, these examples are only for the understanding of the present invention, and the scope of the present invention is not limited by them in any sense.

Example  1: Error analysis of hyperparallel sequencing

Figure 3 is a process flow diagram showing an embodiment of the error analysis method of hyper-parallel sequencing. As shown in FIG. 3, the entire experimental procedure consists of target template preparation, first order hyperparallel sequencing, bead separation with error sequences, barcode PCR, and second order hyperparallel sequencing illumination sequencing and analysis.

1.1. target  Manufacturing of the template

Polymerase error was chosen as the actual mutation that is distinct from the error that occurs in hyperparallel sequencing analysis. Since the error rate of insertion or deletion indel during error due to polymerase is very low at 10 ^-9 / base-cycle level, it is difficult to observe and analyze experimentally, so only mismatch variation is considered as actual variation. . To rule out other variations, an essential gene of E. coli with little mutation was used as a template and 11 pairs of different primers were constructed to amplify the gene. PCR was performed using Phusion DNA polymerase having a mismatch error rate of about 10 ⁻⁶ / base · cycle level so that mismatch errors could be observed in the first super-parallel sequencing analysis.

1 μl colony of E. coli , 2.5 μl each of 10 μM forward and reverse primer, 10 μl 5 × buffer, 1 μl deoxynucleotide, 0.5 μl Phusion® High-Fidelity DNA Polymerase, nuclei 50 μl of the PCR reaction mixture solution containing water was reacted under the following conditions: a cycle consisting of 15 seconds at 98 ° C., 30 seconds at 62 ° C., and 1 minute at 72 ° C. after reaction at 98 ° C. for 30 seconds. After 20 repetitions, the mixture was reacted at 72 ° C for 6 minutes and stored at 4 ° C.

1.2. First Hyperparallel Sequencing Analysis

The first superparallel sequencing analysis was performed on the target template prepared in 1.1 above using the GS Junior 454 sequencing system (Roche).

E. coli published in NCBI among the obtained sequencing data Compared with the base sequence of the gene of interest, the base sequence including insertion, deletion, or substitution was determined as an error sequence. As a result of the first sequencing, a total of 42,328 reads corresponding to the gene dapA2 had 10,878 reads, of which 3,196 reads were detected and the ratio was 29.38%.

1.3. With error sequences Bead  Separation

Beads with error sequences in 1.2 above were separated from sequencing plates via Sniper cloning (Nature communications, 6, Article number: 6073 (2015)). A total of 3,196 beads were separated.

Specifically, the primary superparallel sequencing result data output in FASTA format includes ID of each bead, base sequence information, and coordinate values in the sequencing plate. Using the coordinate values and the image of the sequencing plate, the wells at the desired positions were identified under a microscope, and the lasers were irradiated at the positions to collect the beads in separate plates.

1.4. Separated To DNA  Barcode addition

In order to add one barcode sequence to one isolated bead, PCR was performed using a primer containing the barcode sequence. A total of 20 96-well plates were barcode PCR per well. The forward primer used in the barcode PCR indexes the number of the sequencing plate, and the reverse primer functions to index the well number of the plate.

To each well, 20 μl of a PCR reaction mixture solution containing 0.2 μl of 100 μM forward and reverse primers, 10 μl of Taq DNA polymerase master mix and 9.6 μl of nuclease-free water was added, followed by reaction at 95 ° C. for 2 minutes. After performing 27 cycles consisting of 20 seconds at 95 ° C, 40 seconds at 58 ° C, and 30 seconds at 72 ° C, the reaction was carried out at 72 ° C for 5 minutes and stored at 4 ° C.

4 shows some results of barcode PCR. The numbers listed in each lane in Figure 4 represent the well numbers of the 96 well plates used.

1.5. Secondary Super Parallel Sequencing Analysis

The PCR products barcoded in 1.4. Were pooled and subjected to secondary parallel sequencing analysis. By sorting by barcode, the results of primary and secondary superparallel sequencing analysis on the same DNA sample can be collected. Error sequences that occur during hyperparallel sequencing are more likely to be present at random locations than at certain locations. On the contrary, when the mutant sequence is present on the actual DNA, the PCR amplification product has the same mutant sequence, and thus, the sequencing analysis shows that the mutant sequence is mostly located at the same position. However, since the mutated sequence on the actual DNA may appear differently due to an error by sequencing analysis, in this experiment, when the sequence having the same barcode shows more than 90% of the mutated sequence at the same position, the mutant sequence is identified as actual DNA. The variant sequence present on the sample was considered.

Specifically, 300 bp paired end sequencing was performed using a PEAR tool, a paired end merge tool, on FASTA-type sequence information obtained through Miseq illumina sequencing. Next, the second superparallel sequencing analysis data having the same barcode and the sequence of the corresponding bead binding DNA were aligned using a bead sequence alignment (BWA) tool. After converting the SAM / BAM file format, single nucleotide polymorphism (SNP) analysis was performed using the bam-readcount program.

5 is a visualized representation of the BWA alignment results. A reference is a result of 454 sequencing corresponding to 17_A1, and a query is an alumina sequencing result corresponding to 17_A1. For all barcodes, SNPs were expected to be 2.18 × 10 ⁻⁶ / bp / cycle when more than 90% of the mutations were present at the same location.

Example  2: Error analysis of hyperparallel sequencing

6A and 6B show graphs comparing sequencing results by barcodes. The x-axis and y-axis show the frequency and position of the bases, respectively. In FIG. 6A, 17_A1 means well number A1 of plate 17. FIG. In the first sequencing, the guanine was mutated to thymine at position 191, but the sequencing result was not called at the position. In FIG. 6B, for 17_B8, both primary and secondary sequencing results indicated that there was a sequence of cytosine-mutated guanine at position 119. Therefore, the first sequencing data corresponding to 17_A1 was determined to be an error generated in 454 sequencing, and in the case of 17_B8, it was determined to be an actual mutated sequence.

Claims (21)

(a) selecting at least one first read to be verified among reads generated by analysis through hyperparallel sequencing;
(b) recovering the nucleic acid fragment corresponding to the first read from the sequencing plate using the location information of the nucleic acid fragment;
(c) performing sequencing on each of the fragments or amplification products thereof to generate a second read; And
(d) comparing the second lead with a corresponding first lead to verify an error of the first lead;
At least one of the first reads comprises a sequence mutated based on the reference sequence or comprises the same sequence as the reference sequence Error checking method of hyperparallel sequencing. delete delete delete The method of claim 1, wherein step (b) is by detaching a portion of the support or sequencing plate to which the fragments are bound, through application of energy or by a magnetic material. The method of claim 5, wherein said energy application is via laser irradiation. The method of claim 1, wherein in step (b), each of said fragments is recovered in a space separated state or added with a unique identification material. The method of claim 1, wherein in step (b) the fragments are recovered in a mixed state and a unique identifier is added to each fragment prior to step (c). The method of claim 7 or 8, wherein the substance is a biochemical molecule comprising a nucleic acid, a polypeptide, a peptide, or a combination thereof. The method of claim 1, wherein the hyperparallel sequencing is selected from the group consisting of sequencing by synthesis, ion torrent sequencing, pyro sequencing, sequencing by ligation, nanopore sequencing, and single-molecule real time sequencing. The method of claim 1, wherein the sequencing comprises hyperparallel sequencing, Sanger sequencing, mass spectrometry, electrophoresis, hybridization, digital PCR, allele-specific PCR, quantitative PCR, fluorescence based classification, or a combination thereof. How. The method of claim 1, wherein the step (d) is performed when the ratio of reads having the same sequence as the first reads at a corresponding site with the first reads among the second reads corresponding to the first reads is less than a predetermined value. Determining that site of the first read is due to a sequencing error. A read selector configured to select at least one first read to be verified among reads generated by the analysis through hyperparallel sequencing;
A nucleic acid recovery unit for recovering a nucleic acid fragment corresponding to the first read from a sequencing plate using position information of the nucleic acid fragment;
A lead generation unit generating a second lead through sequencing of each fragment or amplification product thereof; And
A determination unit comparing the second lead with a first lead corresponding thereto and verifying an error of the first lead;
Wherein at least one of the first reads comprises a mutated sequence based on a reference sequence or comprises the same sequence as the reference sequence. delete delete The apparatus of claim 13, wherein the nucleic acid recovery unit comprises a nucleic acid reaction unit that adds a unique identification substance to each fragment through a hybridization reaction or an amplification reaction by PCR. The apparatus of claim 13, wherein the nucleic acid recovery portion comprises an energy applying portion for applying energy to detach a portion of the support or sequencing plate to which the nucleic acid fragment corresponding to the first read is bound from the plate. The apparatus of claim 13, wherein the nucleic acid recovery unit comprises a laser light source, an imaging device, and a light collecting device. The apparatus of claim 13, wherein the nucleic acid recovery unit comprises a storage container for recovering the respective fragments in a space separated form from each other. The method of claim 13, wherein the read generation unit generates the second read by hyperparallel sequencing, Sanger sequencing, mass spectrometry, electrophoresis, hybridization, digital PCR, allele-specific PCR, quantitative PCR, fluorescence based classification, or a combination thereof. Device. The method according to claim 13, wherein the determining unit according to the ratio of the read of the second lead corresponding to the first lead having the same sequence as the first read in the corresponding site with the first lead of the corresponding site of the first read Device for determining whether a sequencing error exists.

Images